BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Neuro-Symbolic Deep Natural Language Understanding - Lili Mou (Uni
 versity of Alberta)
DTSTART:20220517T110000Z
DTEND:20220517T120000Z
UID:TALK174476@talks.cam.ac.uk
CONTACT:Michael Schlichtkrull
DESCRIPTION:Abstract: \n\nDeep learning has largely improved the performan
 ce of natural language understanding (NLU) systems. However\, most deep le
 arning models are black-box machinery\, and lack explicit interpretation. 
 In this talk\, I will introduce our recent progress on neuro-symbolic reas
 oning for NLU\, which combines different schools of AI\, namely\, symbolis
 m and connectionism. Generally\, we will design a neural system with symbo
 lic structures for an NLU task\, and apply reinforcement learning or its r
 elaxation to perform weakly supervised reasoning in the downstream task. O
 ur framework has been successfully applied to various tasks\, including SQ
 L command reasoning\, syntactic structure reasoning\, and logical reasonin
 g. \n\n\nBio: \n\nDr. Lili Mou is an Assistant Professor at the Department
  of Computing Science\, University of Alberta. He is also an Alberta Machi
 ne Intelligence Institute (Amii) Fellow and a Canada CIFAR AI (CCAI) Chair
 . Lili received his BS and PhD degrees in 2012 and 2017\, respectively\, f
 rom School of EECS\, Peking University. After that\, he worked as a postdo
 ctoral fellow at the University of Waterloo. His research interests includ
 e deep learning applied to natural language processing as well as programm
 ing language processing. He has publications at top conferences and journa
 ls\, including AAAI\, EMNLP\, TACL\, ICML\, ICLR\, and NeurIPS. He also pr
 esented tutorials at EMNLP'19 and ACL’20.
LOCATION:Computer Lab\, FW26
END:VEVENT
END:VCALENDAR